Article
renal
Pyelonephritis
William V. Raszka, Jr,
MD,* Omar Khan, MD†
Author Disclosure
Drs Raszka and Khan
did not disclose any
financial relationships
Objectives
After completing this article, readers should be able to:
1. Describe the epidemiology of urinary tract infections and pyelonephritis in children.
2. Discuss the risk factors for the development of pyelonephritis.
3. Compare and contrast methods to diagnose urinary tract infection and pyelonephritis
in children.
4. Describe the management of pyelonephritis.
5. Explain the long-term complications of pyelonephritis.
relevant to this
article.
Introduction
The frequency of urinary tract infections (UTIs) is second only to that of respiratory tract
infections in the pediatric population. UTIs often are separated into infections of the lower
urinary tract that involve the bladder and urethra and those of the upper tract that involve
the kidneys, renal pelvis, and ureters. Infections of the upper tract are designated pyelonephritis. A complicated UTI, whether of the upper or lower tract, usually is associated
with an underlying condition that increases the likelihood of a therapeutic failure.
Underlying conditions include abnormal anatomy, urologic dysfunction, the presence of
an indwelling catheter, or isolation of a multiresistant organism.
Epidemiology
Although the prevalence of UTI has been studied in a variety of patient populations, there
are fewer data regarding the prevalence of pyelonephritis due, in part, to the difficulty in
distinguishing upper from lower tract disease. The prevalence of UTI is influenced by
factors such as age, sex, population sample, urine collection method, testing methodologies, diagnostic criteria, and culture. Age and sex are the most important factors. In
newborns, the prevalence of UTI in preterm infants (2.9%) exceeds that of term infants
(0.7%). UTI is more common in preschool-age children (1% to 3%) than in school-age
children (0.7% to 2.3%). During adolescence, both sexually active girls and homosexual
boys are at increased risk for developing UTI.
Sex has an enormous impact on the prevalence of UTI. In a retrospective populationbased study, the cumulative incidence rate during the first 6 years of life was 6.6% for girls
and 1.8% for boys. In the first 3 postnatal months, UTI is more common in boys and five
to ten times more common in uncircumcised boys than in circumcised boys. Thereafter,
girls are far more likely to develop symptomatic UTI. The prevalence rate of UTI is 1% to
3% in girls 1 to 5 years old and 1% in school-age children. The prevalence rate in school-age
boys is 0.03%.
Although risk factors for the development of pyelonephritis have not been described
well, risk factors for the development of UTI include a previous history of UTI, siblings
who have a history of UTI, female sex (presumably because of the short female urethra), an
indwelling urinary catheter, an intact prepuce in boys, and structural abnormalities of the
kidneys and lower urinary tract. Up to 50% of infants may have an underlying structural or
physiologic abnormality of the urinary tract detected at the time of their first UTI.
Vesicoureteral reflux is the most common and important risk factor for the development of
pyelonephritis.
*Associate Professor of Pediatrics and Director, Infectious Disease Service, Department of Pediatrics, University of Vermont
College of Medicine, Burlington, Vt.
†
Department of Family Medicine, Fletcher Allen Health Care, Burlington, Vt.
364 Pediatrics in Review Vol.26 No.10 October 2005
renal pyelonephritis
Although many Enterobacteriaceae and other organisms may cause UTI in children, Escherichia coli is the
most common pathogen. E coli can be isolated in approximately 90% of patients at the time of their initial UTI
and in more than two thirds of patients who have
recurrent UTI. Other organisms commonly found in
patients who have community-acquired UTI include
Enterobacter, Proteus, and Klebsiella sp. Streptococcus
agalactiae may cause pyelonephritis in neonates. Enterococcus sp may cause up to 5% of UTIs and often is
associated with more complex genitourinary tract abnormalities. Coagulase-negative staphylococci and
Lactobacillus sp are rare causes of cystitis or pyelonephritis.
Pathophysiology
In almost all children outside the immediate newborn
period, pyelonephritis develops after fecal flora colonize
the urethra and ascend into the bladder and kidneys. The
interval between colonization, infection, and disease is
not known. Children can present
with pyelonephritis without preceding symptoms typical of lower
tract infection.
Several virulence factors have
been identified in organisms that
cause pyelonephritis. A small
number of E coli O serotypes are
responsible for approximately
80% of cases of pyelonephritis,
60% of cystitis, and 30% of asymptomatic bacteriuria but are found in only 30% of fecal
flora. The most important virulence factors, particularly
in patients whose urinary tracts are normal, govern adherence or attachment to host mucosal cells. Adhesins,
molecules mediating attachment, are on the surfaces of
bacteria or bacterial appendages. Three major adhesins
associated with strains of E coli causing UTI are PAP,
AFA, and SFA. PAP is the best known and most closely
associated with pyelonephritis. This molecule is located
on the P pili or fimbriae and recognizes P blood group
determinants found on cells lining the urinary tract.
Operons that control expression of PAP, AFA, and SFA
molecules are found in most E coli isolated from patients
who have pyelonephritis.
Host defenses against the development of UTI relate
primarily to anatomic and physiologic considerations.
The ability to empty the bladder of urine regularly and
completely is the most important host defense mechanism against infection. Patients who do not completely
empty the bladder, regardless of the reason, are at signif-
icantly greater risk for the development of a bladder
infection.
The most important risk factor for the development of
pyelonephritis in children is vesicoureteral reflux (VUR).
VUR is detected in approximately 10% to 45% of young
children who have symptomatic UTI. Although there is
some debate about how reflux damages the renal parenchyma, refluxed infected urine from the bladder increases
the risk of pyelonephritis and, in young children, renal
complications such as scarring.
Clinical Findings
The clinical manifestations of pyelonephritis are highly
variable. Older children may present with fever, chills,
nausea, and flank pain in addition to typical signs of lower
tract disease such as dysuria, urgency, and increased
urinary frequency. Clinical findings include suprapubic
and costovertebral (CVA) tenderness. Flank pain, fever,
and vomiting occur more commonly in patients who
have pyelonephritis than in those who have lower tract
Fever
and irritability are the most
common presenting findings in infants who
have pyelonephritis.
disease, but these finding are neither specific nor sensitive
for pyelonephritis. As many as 25% of children who have
no classic signs or symptoms of pyelonephritis are ultimately confirmed to have upper tract disease; up to 50%
of children who have flank pain have no other evidence of
pyelonephritis.
Fever and irritability are the most common presenting
findings in infants who have pyelonephritis. Other findings include poor feeding, lethargy, and in toddlers,
abdominal pain. Traditionally, pyelonephritis is suspected in any infant or child who has fever, emesis, flank
pain, or CVA tenderness on physical examination and a
positive urine culture.
Diagnosis
UTI should be considered in any child 2 months to 2
years of age or younger who has unexplained fever.
Approximately 5% to 7% of febrile infants younger than
8 weeks of age have a UTI compared with approximately
5% of febrile children younger than 1 year of age and 2%
Pediatrics in Review Vol.26 No.10 October 2005 365
renal pyelonephritis
Sensitivity and Specificity
Ranges for Components of the
Urinalysis in the Diagnosis of
Urinary Tract Infection
Table.
Test
Leukocyte esterase
White blood cells on
microscopy
Nitrite
Bacteria on microscopy
Sensitivity
(%)
(range)
Specificity
(%)
(range)
70 to 95
30 to 100
65 to 90
45 to 100
15 to 80
15 to 100
90 to 100
10 to 100
of all febrile children younger than 5 years of age. Patients usually are diagnosed following examination of
their urine. The purpose of the urinalysis is to identify
those children at greater or lesser risk of having a UTI.
Culture of an appropriately obtained urine specimen is
the laboratory test that establishes or excludes the presence of UTI.
Urine can be collected by bag technique, mid-stream
clean catch, urethral catheterization, or suprapubic aspiration. Although many physicians continue to rely on
bagged urine specimens, UTI should not be diagnosed
by a bagged specimen because of its low specificity and
poor positive predictive value. A negative culture from an
appropriately obtained bagged specimen, however, may
be useful in some situations to exclude UTI. For a child 2
months to 2 years of age, urine for culture should be
obtained by suprapubic aspiration or transurethral catheterization. Both of these methods are simple and easy to
perform and give reliable results. In older children, midstream clean catch of spontaneously voided urine after
appropriate cleansing of the urinary meatus is acceptable.
A clean urine specimen can be used to assess for
evidence of inflammation or bacteria. The leukocyte
esterase test identifies an enzyme produced by white
blood cells. A positive leukocyte esterase test usually
corresponds to at least 5 white blood cells per highpower field, with a sensitivity of 70% to 95% (Table).
Direct visualization of the unspun urine gives information about the number and characteristics of white blood
cells. More than 5 white blood cells per high-power field
represents pyuria. Similar to the leukocyte esterase test,
the presence of white blood cells in the urine is a sensitive
but nonspecific test for UTI. The exception is the finding
of white blood cell casts, which, in the absence of signif366 Pediatrics in Review Vol.26 No.10 October 2005
icant hematuria, usually is diagnostic of pyelonephritis.
Some laboratories perform an “enhanced urinalysis,”
which involves examining a drop of urine on a hemocytometer. The presence of more than 10 white blood cells
suggests a UTI.
Many gram-negative organisms reduce nitrates to nitrites in the urine. A positive nitrite test is an insensitive
but relatively specific test for UTI (Table). The presence
of bacteria in a spun urine sample is neither specific nor
sensitive enough to use clinically. Although laboratories
rarely perform routine Gram stains on urine, a positive
Gram stain result on an unspun urine specimen is diagnostic of UTI.
Many children who have radiographically confirmed
pyelonephritis have elevated peripheral white blood cell
counts or nonspecific markers of inflammation such as an
increased erythrocyte sedimentation rate or C-reactive
protein value, but these laboratory tests are neither sensitive nor specific enough to confirm or exclude pyelonephritis.
The diagnosis of UTI is confirmed by the growth of a
significant number of bacteria in a properly obtained
urine specimen. Based on studies of healthy women more
than 50 years ago, growth of more than 105 organisms/
mL of voided midstream urine is indicative of UTI.
Specimens that grow fewer than 104 organisms are unlikely to represent UTI. In the appropriate clinical setting, growth of 104 and 105 organisms may represent
UTI. Any growth of gram-negative bacteria in urine
obtained by suprapubic aspiration indicates a UTI. For
children in whom urine is obtained by transurethral
catheterization, some have proposed that more than
50,000 colonies/mL of urine be used as the criterion for
diagnosing the infection. Urine cultures do not distinguish between lower and upper urinary tract infection
Approximately 10% to 20% of adult women who have
pyelonephritis have concomitant bacteremia. In children, the prevalence of bacteremia in patients younger
than 2 months of age diagnosed as having a UTI is 22.7%
and in patients between the ages of 2 months and 36
months of age is 3.0%. Infants and children whose urine
and blood cultures are positive are assumed to have
pyelonephritis.
Imaging studies are needed infrequently to confirm a
diagnosis of pyelonephritis. Computed tomography
scans are more sensitive than ultrasonography at detecting changes consistent with pyelonephritis, but they
require the use of intravenous contrast. DMSA renal
scintography is the imaging study of choice to confirm or
exclude the diagnosis of pyelonephritis as well as to
detect renal scarring. Its routine use, however, is limited
renal pyelonephritis
to situations such as the challenging patient who has a
neurogenic bladder or significant genitourinary abnormalities.
Treatment
Various antimicrobial agents are useful for the treatment
of UTI. Antimicrobial selection and route of administration are guided by local microbial resistance patterns and
the ability of the patient to take and retain oral agents.
Many commonly used antibiotics are concentrated in the
urine and are useful in treating lower tract disease. Renal
parenchymal antibiotic levels, however, more closely resemble bloodstream levels. The consequence is that nitrofurantoin, which achieves negligible levels in the
blood but is concentrated in the urine, is a good drug for
the treatment or prophylaxis of lower UTI but not for
pyelonephritis.
Children who are suspected of having pyelonephritis
are started empirically on antibiotics effective against the
usual pathogens for their clinical condition. Children
antibiotics had no advantage over oral antibiotics regarding time to defervescence, renal scarring, duration of
hospitalization, and time to urine sterilization.
Many oral agents can be used for the empiric therapy
of pyelonephritis in patients well enough to take oral
medications. Amoxicillin, however, is no longer an acceptable choice because approximately 40% of E coli
strains elaborate a penicillinase. TMP-SMX has been a
standard first-line oral agent for some time and may be
more effective than beta-lactams at treating UTIs. However, E coli resistance to TMP-SMX is becoming more
common. TMP-SMX should not be used for empiric
therapy of pyelonephritis if local resistance to the drug
in E coli exceeds 10% to 20%. Alternatives to TMPSMX include first-, second-, and third-generation oral
cephalosporins, penicillin/beta-lactamase inhibitor
combinations, and for the child older than 12 months of
age, ciprofloxacin. Because of concerns about adverse
effects of ciprofloxacin in young children, when it appears to be the best drug, the potential benefits and
problems associated with it
should be discussed with the parents.
For patients who have a confirmed UTI, antimicrobial susceptibility testing results should
guide therapy. Whenever possible, amoxicillin or TMP-SMX
should be used. Although few
randomized trials comparing
treatment regimens for pyelonephritis in children have been performed, the available
evidence suggests that 10 to 14 days of antimicrobial
therapy is adequate for uncomplicated pyelonephritis.
After completion of 10 to 14 days of antibiotics for their
initial UTI, children ages 2 months to 2 years should be
continued on prophylactic antibiotics until imaging
studies are completed and VUR is excluded.
Follow-up urine culture after 24 hours of therapy is
not indicated because it rarely is positive. Almost 90% of
infants who have febrile UTIs and presumed pyelonephritis were afebrile within 48 hours of initiation of
antibiotics. Those who remained febrile for more than
48 hours did not have any different laboratory or clinical
findings from those who became afebrile in fewer than
48 hours. Most important, there was not a higher rate of
complications in those who failed to respond in 48 hours.
However, many clinicians re-evaluate children who have
not become afebrile within 48 hours with another urine
culture or an imaging study, looking for evidence of
urinary tract obstruction.
For
patients who have a confirmed UTI,
antimicrobial susceptibility testing results
should guide therapy.
deemed toxic or ill enough to require hospitalization
usually are started on intravenous antibiotics. Ampicillin
and gentamicin have been used empirically for decades to
treat pyelonephritis. Gentamicin is effective against most
aerobic gram-negative bacilli; ampicillin is the drug of
choice for enterococci. Alternatives to this regimen include most first-, second-, or third-generation cephalosporins, trimethoprim-sulfamethoxazole (TMP-SMX),
and penicillin/beta-lactamase inhibitor combination
drugs. Parenteral antibiotics most often are continued
until the patient is afebrile for 24 hours.
For children who are presumed to have pyelonephritis
and who appear ill or are vomiting but are not ill enough
to require hospitalization, an intramuscular dose of
ceftriaxone or gentamicin can be administered followed
by oral antibiotics. For children older than 1 month of
age who are suspected of having pyelonephritis, do not
appear toxic, and can take oral agents, oral antimicrobial
agents can be started. At least one study has shown that
in children older than 2 months of age, intravenous
Pediatrics in Review Vol.26 No.10 October 2005 367
renal pyelonephritis
ferred choice for all boys of any age at the time of their
first UTI. Radionuclide cystography often is performed
in girls and for follow-up VCUGs.
Unfortunately, it is uncertain whether the identification and treatment of children who have primary VUR
confer clinically important benefits. In randomized, controlled trials, no treatment or treatment of VUR with
long-term antibiotics, surgery, or a combination of antibiotic prophylaxis and surgery seemed to have similar
results. Natural history studies suggest that long-term
medical management is as effective as surgical therapy for
those who have primary VUR. Currently, children who
have VUR are placed on long-term prophylactic antibiotics such as daily nitrofurantoin, trimethoprim, or
TMP-SMX. Children who have VUR and recurrent UTI
with progressive renal damage despite prophylactic antibiotics or those who have severe reflux often are considered for surgical correction of the VUR.
Complications
Figure. Fluoroscopic VCUG showing grade III reflux.
Prevention
The goal of medical management of patients who have
pyelonephritis not only is to eradicate the infection but to
identify those patients at risk for subsequent infection
and renal scarring. The American Academy of Pediatrics
(AAP) recommends that children younger than age 2
years diagnosed with their first UTI be evaluated for
evidence of urologic abnormalities. Many clinicians also
evaluate older children who have febrile UTIs or pyelonephritis. Although one study suggested that the results
of ultrasonography had no impact on the management of
a select group of children who had an initial UTI, the
AAP recommends that renal ultrasonography be used to
define the anatomy of the urinary tract because the test is
noninvasive, involves no radiation, and can define renal
anatomy readily (eg, obstruction, major scarring, cysts,
or renal dysplasia).
Children who have UTI diagnosed in the first 2 years
after birth should be evaluated with voiding cystourethrography (VCUG). Both fluoroscopic or contrast
VCUG and radionuclide cystography are good at detecting reflux and duplicated systems. Fluoroscopic VCUG
can provide enough information to grade reflux (Figure)
according to an international scale that calculates the
height of the reflux and the associated anatomic changes
to the kidney and identifies bladder and urethral abnormalities better than radionuclide studies. It is the pre368 Pediatrics in Review Vol.26 No.10 October 2005
Patients who have pyelonephritis may develop bacteremia. The likelihood of developing bacteremia depends
on the age of the child. Children younger than 2 months
of age are the most likely to have bacteremia at the time
of presentation. No laboratory test or physical findings
can distinguish between patients who have bacteremia
and those who do not. Children who have bacteremia
have similar clinical courses as those who do not, whether
treated with oral or intravenous antibiotics.
Long-term complications of pyelonephritis include
recurrence, renal scarring, and hypertension. Approximately 18% of boys who have a UTI before age 1 year
develop recurrent infections compared with 32% when
the initial infection happened after the first postnatal
year. Most recurrences occur within the first year of
infection. Similar to boys, girls who have their first UTI
at a later age have a higher recurrence rate than those
diagnosed in the first year (40% versus 26%). Each prior
infection increases the risk of a recurrence by as much as
25%.
Children who have pyelonephritis may develop renal
scarring. For most children who have UTI in whom renal
scars are found, the scars are visible on the first set of
imaging studies. These original scars remain unchanged
irrespective of the child’s future clinical course. Infants
and young children are more prone to scarring than are
children older than age 5 years and adults, although the
reason is unclear. Children treated for UTI who have
evidence of VUR are significantly more likely to develop
scars than are children who do not have VUR. Almost all
new scars in children who have VUR occur before the age
renal pyelonephritis
of 6 years and almost always in association with new
UTIs. More severe reflux is associated with a greater risk
of scarring. A long-term follow-up study of children who
had grade III or IV reflux managed either medically or
surgically found that the incidence of new scar formation
over a 5-year period was 20% in children younger than
age 2 years, 10% in children between 2 and 4 years, and
5% in children older than age 4 years at the time of
enrollment. Only two children developed new scars more
than 5 years after enrollment.
Early detection of pyelonephritis is critical to preservation of renal function. In a recent study of United
States children who were identified as having UTI early
in life, only one child had renal scarring on entry. Other
United States investigators also have found a low incidence of renal scarring in children who had presumed
pyelonephritis identified early in life. This contrasts with
older studies, which demonstrated a high incidence of
scarring.
Long-term follow-up studies have linked renal scarring with hypertension, decreased glomerular filtration,
and end-stage renal disease. Unfortunately, the exact
mechanism of risk of eventually developing hypertension
by patients who have scars is not known. Moreover, it is
unclear why some children who have extensive scars
never develop hypertension. Currently, the best evidence
suggests that children who have pyelonephritis and renal
scarring be followed carefully over time.
Conclusion
Pyelonephritis in the pediatric patient is a serious illness.
Although some aspects of the management of this disease
are still open to debate, there is broad agreement that
prompt recognition and treatment are essential to prevent long-term sequelae.
Suggested Reading
American Academy of Pediatrics. Committee on Quality Improvement. Subcommittee on Urinary Tract Infection. Practice parameter: the diagnosis, treatment, and evaluation of the initial
urinary tract infection in febrile infants and young children.
Pediatrics. 1999;103:843– 852
Chon CH, Lai FC, Shortliffe LM. Pediatric urinary tract infections.
Pediatr Clin North Am. 2001;48:1441–1459
Edelmann CM Jr, Ogwo JE, Fine BP, Martinez AB. The prevalence
of bacteriuria in full-term and premature newborn infants. J Pediatr. 1973;82:125–132
Herr SM, Wald ER, Pitetti RD, Choi SS. Enhanced urinalysis
improves identification of febrile infants ages 60 days and
younger at low risk for serious bacterial illness. Pediatrics. 2001;
108:866 – 871
Hoberman A, Charron M, Hickey RW, Baskin M, Kearney DH,
Wald ER. Imaging studies after a first febrile urinary tract
infection in young children. N Engl J Med. 2003;348:195–202
Keren R, Chan E. A meta-analysis of randomized, controlled trials
comparing short- and long-course antibiotic therapy for urinary
tract infections in children. Pediatrics. 2002;109:e70. Available
at: http://pediatrics.aappublications.org/cgi/content/full/
109/5/e70
Kraus SJ. Genitourinary imaging in children. Pediatr Clin North
Am. 2001;48:1381–1424
Le Bouguenec C, Lalioui L, du Merle L, et al. Characterization of
AfaE adhesins produced by extraintestinal and intestinal human
Escherichia coli isolates: PCR assays for detection of Afa adhesins
that do or do not recognize Dr blood group antigens. J Clin
Microbiol. 2001;39:1738 –1745
Olbing H, Smellie JM, Jodal U, Lax H. New renal scars in children
with severe VUR: a 10-year study of randomized treatment.
Pediatr Nephrol. 2003;11:1128 –1131
Pitetti RD, Choi S. Utility of blood cultures in febrile children with
UTI. Am J Emerg Med. 2002;20:271–274
Ronald A. The etiology of urinary tract infection: traditional and
emerging pathogens. Am J Med. 2002;113(suppl 1A):14S–19S
Wheeler DM, Vimalachandra D, Hodson EM, Roy LP, Smith GH,
Craig JC. Interventions for primary vesicoureteric reflux.
Cochrane Database Syst Rev. 2004:CD001532
Wiswell T. The prepuce, urinary tract infections, and the consequences. Pediatrics. 2000;105:860 – 862
Multimedia Links
AAP Guidelines on the diagnosis, treatment, and evaluation of the
initial urinary tract infection in febrile infants and young children. Available at: http://www.aap.org/policy/ac9830.htm
The NIH/NIDDK Web site on pyelonephritis. Available at: http://
www.niddk.nih.gov/health/kidney/summary/pyelonep/
pyelonep.htm
MEDLINEplus medical encyclopedia on pyelonephritis. Available
at: http://www.nlm.nih.gov/medlineplus/ency/article/000522.
htm
The Atlas of Renal Pathology. Available at: http://ajkd.
wbsaunders.com/atlas/35/2/atlas35_2.htm
Pediatrics in Review Vol.26 No.10 October 2005 369
renal pyelonephritis
PIR Quiz
Quiz also available online at www.pedsinreview.org.
9. Which of the following findings or conditions increases the risk of developing a urinary tract infection
(UTI)?
A.
B.
C.
D.
E.
Acidification of the urine.
Dilution of the urine.
High specific gravity.
Incomplete bladder emptying.
Large bladder capacity.
10. An 18-month-old boy presents with fever for 3 days. Urine is obtained by catheterization. Which of the
following findings in the urine confirms the diagnosis of UTI?
A.
B.
C.
D.
E.
Positive Gram stain of unspun urine.
Positive leukocyte esterase test.
Positive nitrite test.
Presence of bacteria in spun urine.
Presence of 10 white blood cells in unspun urine.
11. A 15-month-old girl who has recurrent UTIs presents with fever and irritability. Analysis of her urine
shows more than 50 white blood cells per high-power field. Which of the following oral medications is
least likely to be effective for the empiric treatment of her presumed UTI?
A.
B.
C.
D.
E.
Amoxicillin.
Cefuroxime.
Cephalexin.
Ciprofloxacin.
Trimethoprim-sulfamethoxazole.
12. A 20-month-old patient is diagnosed as having her first UTI. After 24 hours of antibiotics, she is afebrile.
Renal ultrasonography findings are normal. Which of the following tests should be performed on this girl?
A.
B.
C.
D.
E.
DMSA renal scintography.
Immediate repeat urine culture.
Intravenous pyelography.
Repeat renal ultrasonography in 6 weeks.
Voiding cystourethrography.
370 Pediatrics in Review Vol.26 No.10 October 2005